Analysis of Point Cloud Data Using Depth and Texture Maps

1. A method of analyzing data points in a point cloud system, comprising: receiving three dimensional (3D) point cloud data points; transforming the received 3D point cloud data points to a two dimensional (2D) depth map or a texture map; determining a neighbor data point that is nearest to each of said transformed 3D point cloud data points from a group of said transformed 3D point cloud data points within a bounded window of said 2D depth map or said texture map; and estimating trajectory equations utilizing said neighbor data point that is nearest to each of said transformed 3D point cloud data points and said received 3D point cloud data points.

2. The method as recited in claim 1 , further comprising: storing said 2D depth map; and wherein said estimating utilizes said stored 2D depth map.

3. The method as recited in claim 1 , further comprising: storing said texture map; and wherein said estimating utilizes said stored texture map.

4. The method as recited in claim 1 , further comprising: utilizing a LiDAR system to generate, at a periodic interval, slices of said received 3D point cloud data points; and wherein said 3D point cloud data points represent part of a geometric space of a vehicle.

5. The method as recited in claim 4 , wherein said vehicle is an autonomous or semi-autonomous vehicle.

6. The method as recited in claim 4 , further comprising: utilizing said slices; and wherein said determining, and said estimating are executed for each of said slices.

7. The method as recited in claim 6 , further comprising: storing each of said slices as a portion of said 2D depth map; and utilizing, for each of said slices, a first angle parameter for a horizontal dimension of said each slice, and a second angle parameter for a vertical dimension of said each slice, wherein said first angle parameter is equivalent for each of said slices analyzed and said second angle parameter is equivalent for each of said slices analyzed.

8. The method as recited in claim 1 , wherein said determining said neighbor data point is performed on a graphics processing unit (GPU).

9. The method as recited in claim 1 , wherein said bounded window includes a fixed number of said neighbor data points.

10. The method as recited in claim 9 , wherein said bounded window is 13 pixels wide by 7 pixels high.

11. The method as recited in claim 9 , wherein said bounded window is 9 pixels wide and 5 pixels high.

12. The method as recited in claim 9 , wherein a separate GPU processing thread is spawned for each of said transformed 3D point cloud data points within said bounded window, and where some of said processing threads share fetched data.

13. The method as recited in claim 1 , further comprising: calculating moving object parameters utilizing said trajectory equations.

14. The method as recited in claim 1 , further comprising: generating, utilizing said trajectory equations, at least one geometric space parameter for at least one of obstacle detection, freespace detection, and landmark detection.

15. A point cloud data analysis system to generate geometric space parameters, comprising: a receiver, operable to receive three dimensional (3D) point cloud data points representative of part of a geometric space surrounding a vehicle; and a graphics processing unit (GPU) operable to transform said received 3D point cloud data points to at least one of a two dimensional (2D) depth map and a texture map, to determine a neighbor data point that is nearest to each of said transformed 3D point cloud data points from a group of said transformed 3D point cloud data points within a bounded window of said 2D depth map or said texture map, and to generate said geometric space parameters.

16. The system as recited in claim 15 , further comprising: an initiator, operable to initiate a generation of said 3D point cloud data points at a periodic interval.

17. The system as recited in claim 16 , wherein said initiator is a LiDAR system.

18. The system as recited in claim 15 , wherein said vehicle is one of an autonomous vehicle and a semi-autonomous vehicle.

19. The system as recited in claim 15 , wherein said geometric space parameters are associated with at least one of a moving object, obstacle detection, freespace detection, and landmark detection.

20. A computer program product having a series of operating instructions stored on a non-transitory computer-readable medium that directs a data processing apparatus, when executed thereby, to perform operations comprising: receiving three dimensional (3D) point cloud data points, at a periodic interval, representative of part of a geometric space surrounding a vehicle; transforming said received 3D point cloud data points to a two dimensional (2D) depth map or a texture map; determining a neighbor data point that is nearest to each of said transformed 3D point cloud data points from a group of said transformed 3D point cloud data points within a bounded window of said 2D depth map or said texture map; and estimating trajectory equations for said vehicle utilizing said neighbor data point that is nearest to each of said transformed 3D point cloud data points.

21. The computer program product as recited in claim 20 , wherein said determining is performed utilizing a graphics processing unit (GPU).

22. The computer program product as recited in claim 20 , wherein said bounded window includes a fixed number of said neighbor data points.

23. The computer program product as recited in claim 20 , further comprising: generating, utilizing said trajectory equations, at least one geometric space parameter for at least one of moving object detection, obstacle detection, freespace detection, and landmark detection.

24. The computer program product as recited in claim 20 , further comprising: utilizing said periodic intervals; and wherein said receiving, said determining, and said estimating, are executed for each of said periodic intervals.

25. A method of storing three dimensional (3D) sensor ranging information, comprising: collecting 3D point cloud data points employing a detection and ranging sensor; storing said collected 3D point cloud data points as a two dimensional (2D) depth map; and determining a neighbor data point that is nearest to each of said stored 3D point cloud data points from a group of said stored 3D point cloud data points within a bounded window of said 2D depth map.